As expected, Nvidia has now officially announced its newest Geforce GTX 980 graphics card based on the GM204 Maxwell GPU. Packing quite a few new features thanks to the new second-generation Maxwell GPU as well as put a lot of pressure on the competition as it ends up anywhere between 2 to 36 percent faster than the Radeon R9 290X, depending on the game.


As noted in earlier reports, the new Geforce GTX 980 is based on a fully enabled GM204 Maxwell GPU. Based on a 28nm manufacturing process and featuring a staggering 5.2 billion transistors on a 398 mm² die size. Just as it was the case with Fermi and Kepler GPU, the GM204 is also composed of an array of Graphics Processing Clusters (GPCs), or four in this case. In the Geforce GTX 980, each GPC features a dedicated raster engine and four Streaming Maxwell Multiprocessors (SMMs). This adds up 16 SMMs, each with 128 CUDA cores, a PolyMorph Engine and eight texture units, adding up to 2048 CUDA cores and 128 Texture Units (TMUs).




Each GPC also features a memory controller, or in this case four 64-bit memory controllers for 256-bit wide memory interface. Each memory controller is connected to 16 ROP units and 512KB of L2 cache, which adds up to a total of 2048KB of L2 cache and 64 ROPs, which is quite an improvement compared to the the Kepler GK104 GPU which features 512KB of L2 cache and 32 ROPs, as well as had half the amount of SMMs.




The memory subsystem on the GM204 has also been reworked in order to reduce DRAM bandwidth demands as it now use lossless compression techniques as data is written out to memory. With multiple layers of compression algorithms as well as certain caching improvements, Nvidia claims that the GPU is able to significantly reduce the number of bytes that have to be fetched from memory per frame, up to 25 percent less bytes compared to the Kepler GPU.

The reference GTX 980 graphics card works at 1126MHz for the GPU base clock and 1216MHz for the GPU Boost clock. The Geforce GTX 980 also packs 4096MB of GDDR5 memory clocked at 7000MHz, which paired up with a 256-bit memory interface add up to a total memory bandwidth of 224GB/s.




What makes the GM204 Maxwell based Geforce GTX 980 quite interesting is its power efficiency, since it only needs two 6-pin PCI-Express power connectors and has a TDP of 165W.

In order to cool down the GM204 GPU on the Geforce GTX 980, which was not a big feat considering the power efficiency and low 165W TDP, Nvidia decided to use a dual-slot blower-style cooler, similar to the Geforce GTX 780 Ti. It features an aluminum heatsink with three copper heatpipes cooled by a single blower style fan.




What also makes the Geforce GTX 980 interesting is the fact that this is the first reference Geforce graphics card that will feature a backplate, or to be precise, a partially removable backplate. In case you have been wondering about that small piece of the backplate seen removed on some earlier leaked pictures, you might be happy to know that it is there for a reason.




According to Nvidia, a section on the end of the backplate can be removed in order to improve the airflow when the GTX 980 is used in SLI multi-GPU configurations. Apparently, that section of the board is critical for feeding air directly to the adjacent fan.




In addition to the cooler, the Geforce GTX 980 also brings a new I/O bracket that has been stamped with triangles in order to improve the exhaust. Another novelty are the five display outputs (four of which can be used at the same time), including three DisplayPort connectors, HDMI 2.0 connector and a dual-link DVI connector. This allows the Geforce GTX 980 to drive three G-SYNC enabled displays and it feature HDMI 2.0 output that supports 4K@60Hz output.




The new Nvidia Geforce GTX 980, or to be precise, the GM204 Maxwell GPU, will also bring some new features including the Voxel Global Illumination (VXGI) lighting, Dynamic Super Resolution, Multi-Frame sampled Anti-Aliasing (MFAA) and a new Nvidia VR Direct Technology. The VXGI lighting uses a voxel grid and cone tracing technique to calculate an approximation of global illumination as well as ambient occlusion on the GPU in real time, includes support for a number of new graphics features that dramatically speed up the voxelization process where we convert triangles into voxels and dramatically speeds up the Global Illumination (GI). The VXGI will be available for Unreal Engine 4 and other major game engines in Q4 2014.




The next in line is the Dynamic Super Resolution, which Nvidia describes as a new technology that is designed to provide gamers with better looking games. It pracitcally renders the game at higher resolution and then scales the image down to the native resolution, thus providing better level of detail.




For those looking for higher image quality at lower frame rates, the GM204 brings a new anti-aliasing method, the Multi-Frame sampled AA (MFAA). The MFAA alternates between multiple AA sample patterns to produce the best image quality while still offering a performance advantage compared to traditional MSAA.




The last but not the least is the new Nvidia VR Direct Technology. With the rise of popularity of virtual reality, thanks mostly to the Oculus Rift, Nvidia has developed the VR Direct Tehnology that should reduce latency as much as possible, thanks to certain latency optimization like the MFAA and asynchronous warp.




We will not go into details regarding performance figures since you can check out our published reviews for more details, but, according to Nvidia, the Geforce GTX 980 ends up anywhere between 2 to 36 percent faster than the AMD Radeon R9 290X, depending on the game.

What is most important is that the today Geforce GTX 980 launch is actually a hard launch, which means that you should be able to find the Geforce GTX 980 on retail/e-tail shelves as of today. The price in Europe is set at €452 ex. tax, while the USA price is set at US $549 ex. VAT.



Source: Nvidia.com.